I have occasionally run across the weird and wonderful word zhuzh /ʒʊʒ/; it turns out the OED has entries for both noun (“Style, glamour; a stylish or glamorous appearance or effect”) and verb (“To smarten up (someone or something); to make (something) more stylish, attractive, or exciting”), both updated just last December. The latter has the full etymology, which is intriguing:

Etymology: Of uncertain origin. Perhaps ultimately an expressive formation (compare e.g. whoosh v., swish adj., and use in similar senses and constructions of e.g. zing v. and zap v.).

Compare from a similar date the related zhuzh n., zhuzhy adj., although the relative priority of the three words is unclear.

The etymology of this word, and of the related noun and adjective, is uncertain and disputed; a number of suggestions have been made, but none of them is entirely unproblematic or confirmed beyond doubt by the available evidence.

Derivation from a Romani verb in the sense ‘to clean’ or a corresponding adjective has been proposed (compare e.g. Hungarian Romani shuzo, British Romani yuzho clean, pure), but this presents both phonological and semantic difficulties, and supporting evidence appears to be lacking.

Dict. South African Eng. connects the present word and the corresponding noun and adjective with South African English slang terms for ‘excellent, smart, attractive’ derived from regional pronunciations of Jewish adj., which appear to have been motivated by the high reputation of Jewish tailors and tailoring (perhaps compare quot. 1968 at zhuzh n., apparently in the sense ‘clothing’). However, this is difficult to reconcile with the earliest documentation for zhuzh v. and related words, which suggests a British context, and no evidence has been found for a corresponding sense or pronunciation of Jewish adj. outside South Africa.

That early noun citation is:

1968 B. Took et al. in B. Took & M. Coward Best of ‘Round the Horne’ (2000) 4th Ser. Episode 4. 192/1 Julian. Let’s have a vada at his zhush. Mr. Horne. Clothing. That’s translator’s note.

The first verb cite is slightly later:

1970 P. Burton Lang. of their own: Polari, West End Homosexual Slang (typescript) (O.E.D. Archive) p. ii A zhooshy quean is a grand quean, to zhoosh up is to get ready.

As for the possible Romani derivation, my Цыганско-русский и русско-цыганский словарь (кэлдэрарский диалект) has vužó for ‘clean,’ which just goes to complicate things. And I posted on Polari way back in 2003.


  1. If Russian word zhuzh existed, it would have meant “a single instance of a buzzing sound”. In reality, it is zhuzhzhanie which means “buzz”. But I like it. “The mosquitoes were close by, but for now only occasional zhuzh could be heard”

  2. I didn’t realize it was that old. I’ve only heard it fairly recently on TV shows such as Queer Eye for the Straight Guy, which is consistent with the link to Polari.

  3. The RomLex site offers Romani variants even closer to English zhuzh with zh- than the OED’s examples, such as žužerel, džužerel, and južerel “to clean, cleanse, sweep”, for Welsh Romani:


    And žužo “clean” is also found in Slovak Romani, too:


    If these links don’t work and you want to see the forms, go to the homepage


    and click on “Lexical Database” in the top right, then first select the gloss language (such as English) and then select the Romani variety you want to see, and type in clean or cleanse or sweep or the like in the search box. I wish the site provided bibliographic sources or notes on the collection of forms from field work for these entries, however.

    Turner’s etymology for the Romani word is here, under no. 2448, Old Indic r̥jú- “straight, honest”:


    A u-stem adjective from PIE *h₃reǵ- “straight, straighten, move in straight line, right oneself”, the same formation seen in Vedic urú- “wide” (beside váras- “expanse, room, space”), pr̥thú- “broad” (beside práthate “becomes wider, extends, spreads, increases”), mr̥dú- “soft” (beside mradayati, “softens”), etc.

    On the status of the phoneme /ʒ/ in Romani, this is of note from Y. Matras (2002), Romani: A Linguistic Introduction, p. 51:

    Both /h/ and /ž/ are marginal in the pre-European component [of the lexicon]; the latter may even be regarded as rare (užo ‘clean’ < [Middle Indo Aryan] ujju being an isolated example)

    I want the Romani etymology to be true, just for the piquancy of having the most well-known Polari word come from the Old Indic word for “straight”.

  4. I first heard the word about a month ago from Australian comedian Sarah Kendall, who talked about trying to “zhuzh it up” on the British TV show Taskmaster. I remember being taken aback to hear a word in English both starting and ending with /ʒ/.

  5. William Boyd says

    Whenever I come across such oddities, I wonder, “Scrabble-legal?” I’ll take that up with my better 2/3rds.

  6. @William Boyd, apparently neither “zhuzh” nor “zhoozh” is Scrabble-legal per the online resources I checked for the NASPA Word List nor Collins Scrabble Words.

  7. Xerîb: You’ve outdone yourself! Since that disposes of the OED’s “phonological difficulties,” I hereby declare Romani the winner and the piquancy proved. And if it’s not a Scrabble word, it should be.

  8. Both /h/ and /ž/ are marginal in the pre-European component [of the lexicon]; the latter may even be regarded as rare

    My stupid question: how doesn’t “marginal” imply “rare”?

  9. Seeing as word-initial “th” in any new English word would be pronounced /θ/, would it be fair to call word-initial /ð/ in English “marginal”? But it’s not rare.

    Or does “not rare” indeed imply “not marginal”?

  10. John Cowan says

    A marginal phoneme is one that appears only in marginal words: interjections, unassimilated loanwords, onomatopoeia, etc., and not in the ordinary words of the language (or only in a small number of them, like the emphatic “l” in Arabic Allah, which is literally the only word in the language to have it). The zh-sound was marginal in English until the third palatalization, when people started to say “vizhon” instead of “vizyon”, and is still marginal in German. But that does not make it rare: German words like Garage (the second “g”) and Genie are not rare words.

    rosie: Indeed, any new English word with /ð/ would be extraordinary: it almost always arrives as /d/, as in skordalia, where it is phonemic in Greek, or padre, where it is not phonemic in Spanish.

  11. David Marjanović says

    and is still marginal in German. But that does not make it rare:

    Depends on the speaker (and partially on where they’re from). Many simply can’t pronounce [ʒ] and end up saying /ʃ/. In other words, some people have a loanword phoneme /ʒ/, and some don’t.

    The twist is that many southern accents distinguish a very common /ʃː/ from a rather rare and marginal /ʃ/, and loans with original [ʒ] end up contributing to this rare /ʃ/. But that does not work at the beginnings of words. Except probably in Switzerland.

  12. Trond Engen says

    Davis M.: The twist is that many southern accents distinguish a very common /ʃː/ from a rather rare and marginal /ʃ/, and loans with original [ʒ] end up contributing to this rare /ʃ/

    Would that be because voicing is perceived as one of the available distinguishing features of a short sibillant?

    /ʃː/ [ʃː] | /ʃ/ *[ʒ]
    /sː/ [sː] | /s/ [z]

  13. David Marjanović says

    Not in actual Upper German, where voiced obstruents aren’t a thing even in the most inviting phonetic environments. But the short fricatives can get quite lenis, and then they differ from foreign voiced ones really just by voice alone, phonetically, while the long ones differ by voice + fortis + length.

    Still, this hasn’t happened the other way around. The Carinthian sound system was wholly reinterpreted in Slovene terms, entailing the loss of consonant length. Slovene has of course a /s/ and a /z/. The German /sː/ and /s/ were both interpreted as /s/, so they merged, and there is no /z/ and no [z] in Carinthian – and no [ʒ] either. The plosive system, in contrast, was reinterpreted in terms of voice, so /b d g/ are voiced in Carinthian (and I think there’s even word-final devoicing).

  14. Andrew Dunbar says

    I was really looking forward to reading an article about the Darija word for “two”!

    I don’t think I’ve noticed the English “zhuzh” but I could’ve heard it and just taken it to be a throw-away bit of sound symbolism, maybe like /fʊf/ (never seen it written) as in “to fʊf up” a cushion or one’s hair, to make fluffy or add body. My brain is telling me there might be another word close to the “get dressed up” meaning too, but I can’t think it up.

  15. If it is of any relevance, Chris Denning’s “Polari — British gay slang” lists shush bag ‘holdall’ (https://www.google.com/search?client=firefox-b-1-e&q=polari+vocabulary).

  16. i’ve usually heard “zhuzh” asserted to have a yiddish origin, but haven’t been persuaded, since i haven’t seen anything like it used or lexicographized for anything* in yiddish except “buzz” (cognate with the russian, presumably).

    but since it definitely entered u.s. english in any wide usage through the gay-themed television shows of the last 10-15 years (i think Queer Eye in particular, and then possibly RuPaul’s Drag Race), a polari source makes good sense to me. and all the more so with Xerîb’s confirmation of a plausible rromani origin point!

    i’m not sure i’ve ever run into the word as a noun, though, except in phrases like “give it a zhuzh”.

    * okay, i do find זוזע for “gravy; semen” and זהוזהע for “fire (baby talk)” in refoyl’s database, but those hardly seem relevant.

  17. David Marjanović says

    FOOF, so named because it makes things go FOOF.

    (foof things i won’t work with is a Google suggestion, BTW.)

  18. D. Gold says

    @ rozele

    You are right not to be persuaded about the alleged Yidish origin of zhuzh.

    With exceptions, one should never believe an etymology involving Yidish that one hears on the street, on radio or television, in a newspaper or magazine, at a cocktail party or a picnic, in a bus or taxi, on a train or airplane, or from people claiming to have found “hundreds” of words in English that come from Yidish.

    Also, if one hears or reads an etymology that’s expressed in twenty-five words or fewer, one should demand to see the detailed research backing it.

    (It would be good to see the research supporting a Yidish etymology for zhuzh.)

    Of the thirty-one chapters in the book mentioned below, eight consist of disproofs or are skeptical:

    4. American English Slang copacetic ‘fine, all right’ Has No Hebrew, Yiddish, or Other Jewish Connection

    5. The American English Slangism fink Probably Has No Jewish Connection

    7. Originally American English glitz, glitz up, and glitzy Probably Have No Yiddish Connection

    15. An Immediate or Non-Immediate Jewish Connection for Dutch poeha and Variants (> Afrikaans bohaai > South African English bohaai), French brouhaha (> English brouhaha), French Brou, brou, ha, ha, Brou, ha, ha, High German buhai and Variants, Low German buhê and Variants, or Modern West Frisian bahey and Variants Has Not Been Proven (With Remarks on the Jewish Italian or Liturgical Hebrew Origin of Arezzo Dialectal barruccaba and the Liturgical Hebrew Origin of Italian badanai)

    20. The Etymology of English spiel and spieler and Scots English bonspiel

    (The title should have said that the words have no Yidish connection.)

    21. English Star Chamber Has No Jewish Connection

    25. A Few English Words Misattributed to Yiddish (finagle, finical, finick, toco, trantle, and trantlum); a Yiddish-Origin English Word Misetymologized for at Least Sixty-One Years (bopkes); a Misetymologized Yiddish Pen Name (shmul niger); and a Misetymologized Eastern Yiddish Word (yavne-veyasne!)

    31. Jewish Dickensiana, Part One: Despite Popular Belief, the Name Fagin in Charles Dickens’s Oliver Twist Has No Jewish Connection (With Appendixes on Some Laws Concerning Personal Names and on Dickens’s Authentic Yiddish Name)

    Also pertinent:

    11. Nine Criteria for Assessing the Likelihood of Yiddish Influence on English (With Examples)

    Gold, David L. 2009. Studies in Etymology and Etiology (With Emphasis on Germanic, Jewish, Romance, and Slavic Languages) / Selected and Edited, with a Foreword, by Félix Rodríguez González and Antonio Lillo Buades. Alicante. Publicaciones de la Universidad de Alicante. 870 pages. ISBN 978-84-7908-517-9 [much is available without a paywall at https://books.google.com/books/about/Studies_in_Etymology_and_Etiology.html?id=l015C5vm1XkC%5D.

  19. Trond Engen says

    rozele: okay, i do find זוזע for “gravy; semen”

    Surely, this is the sister of German Soße, from French sauce.

  20. PlasticPaddy says

    Re DGs book:
    Works better than the provided link for me.
    Re zuze, I was puzzled as well. There are several roots with the right onset, but these end up looking like saug- in German or jukh/ukh in Slavic. So I was thinking of German Sud, English suds…

  21. David L. Gold says

    On principle, one should be skeptical of any etymology that posited “British English fashion term < Romani" because of the probable absence of contact between members of the British English fashion world and speakers of Romani.

    Rather, "British English fashion term < Polari < Romani" would be the better bet given the significant presence of gay people in the British fashion world.

    Even if so, one would still have to show that the suggested etymology was phonologically plausible ("if x looks or sounds like y, x must be derived from y" is not a tenet of professionally practiced etymological research; look-alikes and sound-alikes need to be investigated).

    The proposed etymology would also have to be semantically plausible: is meaning 'x' l(of the proposed etymon) likely to become meaning 'y' (of the proposed reflex)?

    In light of the decline of Polari, one would also have to show that zhuzh arose when Polari was still widely used.

    In sum, form, meaning, and time are iindispensable considerations.

    Does the suggested Polari etymology meet those three requirements?

  22. Since the first verb cite is from Polari, and žužo is found in Slovak Romani, the only question is that of semantics, and “To smarten up (someone or something); to make (something) more stylish, attractive, or exciting” seems a plausible development of ‘to clean’ to me. Your mileage may etc.

  23. David L. Gold says

    @LH “Since the first verb cite… ” I agree with you, but it would be good to know why OED is skeptical.

  24. Well, for one thing they didn’t have the benefit of Xerîb’s research.

  25. Andrej Bjelaković says

    Not sure if it’s at all relevant, but I’ve seen Žuža as a nickname for Suzana round these parts.

  26. David Marjanović says

    Straight from Hungarian, I would guess (Zsuzsana).

  27. And why did the Hungarians zhuzh up Susana?

  28. David Eddyshaw says

    Same reason as they zsazsa’d Sári. It is their Hungarian Way, and it is not for us to cavil.

  29. David Marjanović says

    I can offer some waffling about Middle High German… voiced fricatives (at least in the west), apical sibilants that were borrowed as shibilants in Hungarian.

  30. The history of the kind of form represent by the Welsh Romani žužerel, glossed as “clean, cleanse, sweep” by RomLex, is more complicated than I realized.

    Turner indicates that Common Romani appears to have inherited a *šučo, or the like various represented in Greek Romani as šučó, šuzo ʻcleanʼ, Hungarian Romani šužo, British Romani ǰuzo, German Romani čučo, ǰuǰo. These descend from a derivative of the Old Indic root śuc-, exemplified for instance by Vedic śúci- “shining, gleaming, unsullied, undefiled”.


    For the semantic development, note Latin nitidus and French net (English neat). Turner notes that this word collides with the reflexes of the Old Indic word represented by Vedic tucchyá- “empty, vain” (root-related with Lithuanian tuščias, Russian тощий)—hence the overlap of “clean” of “sweep” in Welsh Romani? The entry in Turner is here, number 5850:


    I would like to know whether the identical glosses for the forms given in RomLex indicate that žužerel and južerel are considered variants of the same word by speakers, or they are kept apart. One could see how an original *šučo “clean” and *užo “clean” (from Old Indic r̥jú-) might have collided as well, given the variation among the forms Turner provides.

    It is instructive to look at the RomLex word lists for a variety like Burgenland Romani:

    čučo adj m 1. empty 2. hollow 3. deserted
    šučo adj m empty
    šušo adj m clean, cleanly

    at the following link:


    In any case, the ambiguity of ulterior etymology of the žuž- in forms like Welsh Romani žužerel doesn’t affect its viability as a proximate etymon for English zhuzh, zhush, if we wish to argue for that. I tried to locate the English Romani (Romanichal) forms for “clean” but I was unable to for the moment (it’s hard for me because I can’t type because of an injury). I hope someone else can ferret them out.

    There is another aspect of the family of the word užo that may have bearing on the probability of its adoption into Polari: its importance as designating one pole of the dichotomy of “pure, clean” užo, on the one hand, and måxado (Welsh Romani form) “impure, defiled”, on the other. There is a nice short essay on this aspect of Romani society here:


    Hence žužo might have been one of the first words an English gadjo working in contact with Romani society would have learned?

  31. And why did the Hungarians zhuzh…

    This is not guaranteed to be the case here, but Hungarian zs from an older *s quite often comes thru German, e.g. zsák ‘sack’, zsemle ‘bun’, zsold ‘wage’; maybe zsinagóga from some variety of Yiddish? It’s not clear to me if they’ve been acquired from dialect(s) where not just *s / _C but also plain *s- > *z- is retracted further to postalveolar, or if they merely reflect s having been apical at the time as opposed to z from *t (the “Luigi mechanism”).

  32. David Marjanović says

    It’s pretty clear that the main descendant of Germanic *s was apical in German until the collapse (around the end of MHG) of the three-way distinction between the apical *s, the laminal outcome of *t, and the new postalveolar /ʃ/ from *sk and from *s before other consonants.

    It’s still apical in Dutch, where the other two sibilants never happened.

    Hungarian, in contrast, had a sibilant-shibilant distinction “since ever”, so I expect the sibilants to have been laminal “since ever”, as it is today, in order to maximize the distinction phonetically. Laminal sibilants seem to be pretty much limited to languages that have at least one more place of articulation to contrast them with.

    The one thing I can’t make sense of is the long vowel in zsák.

  33. David L. Gold says

    “maybe zsinagóga from some variety of Yiddish?”

    Yidish is irrelevant to the foregoing discussion about German and Hungarian in general and about the etymology of Hungarian zsinagóga in particular.

  34. Stu Clayton says

    a sibilant-shibilant distinction “since ever”

    That’s the everyday schon immer. What you’re going for, I suspect, is immer schon:

    Das Selbst aber ist zunächst und zumeist uneigentlich, das Man-selbst. Das In-der-Welt-sein ist immer schon verfallen.

    I suggest “forever and a day”.

    That zunächst und zumeist is a tough one, though. “The closest and the mostest”, perhaps ? Older Americans will be reminded of that song in Call Me Madam.

  35. That zunächst und zumeist is a tough one, though. “The closest and the mostest”, perhaps ?
    Literally, yes. But I’d translate it “first of all and most of all”.

  36. Stu Clayton says

    It’s the trite “first and foremost” used only by mayors and other aggrandizing speakers on small occasions. Heidegger loved a Verarsche whenever he could get it in. Generations of commentators and translators have fallen for it.

Speak Your Mind